HTC留学生代写、代做Python程序语言、代写Computing software、Python代做

日期：2018-10-24 09:58

Goal:Write a program to report on failures that are recorded in a complex log file.

Background:

This problem is about a High Throughput Computing software, HTC. But you do not need to know HTC, other than what is described on this page.

HTC runs non-interactive programs, called jobs, for a user on a pool of machines that it manages for that purpose. Often, there are more jobs waiting to run than there are available machines at that moment, and so HTC continuously tracks waiting jobs and available machines and seeks the best way to match one to the other. A few other interesting points:

If a machine fails to run a job,HTC will keep trying to find a machine that can run the job successfully.

The log file for this problem contains information for over 300 jobs, which HTC ran over the course of several hours as machines became available.

HTC recorded many events for each job in the log, most of which you will ignore for this problem.

For this problem, we are interested in reporting on jobs that failed one or more times before completing successfully. This information might be helpful for improving HTC and the pool of machines that it manages.

Failures in the Log File

To analyze the log file, your program must look for failures. Here are two examples of failures:

007 (69180.000.000) 04/16 04:29:49 Shadow exception!

       FAILED TO SEND INITIAL KEEP ALIVE TO OUR PARENT <128.105.244.224:9618?sock=3202_995c_3>

       0  -  Run Bytes Sent By Job

       0  -  Run Bytes Received By Job

...

007 (69188.000.000) 04/16 04:32:07 Shadow exception!

       Error from slot1@c159.chtc.wisc.edu: Failed to create a new VM

       0  -  Run Bytes Sent By Job

       983095  -  Run Bytes Received By Job

...

A few notes about the failures:

Every failure starts with 007 in the first column, as highlighted above

The first number in parentheses (e.g.,69180 or 69188 above) is called the job ID and uniquely identifies a job

A job can fail more than once; each failure will start with 007 and the job ID

The date (month / day) and time of the failure are recorded

After each 007 line, the next line contains the failure message itself

The file contains 007 failures with different messages, too, but they all follow this format, and your program should handle all failure messages. Also, the file contains many other kinds of events and data, which your program should ignore.

How to Analyze the Failures

Write a program to analyze all the failure events as described above. For each unique failure message (the second line of text), tally and output the following pieces of information:

The failure message itself, without leading or trailing whitespace

The number of times that the error occurred

The earliest and latest times that the error occurred; hour and minute are fine, or you may include seconds; also you may assume that all events happened on the same day if you like

The number of unique jobs that were affected; use the job ID in each failure event to determine unique jobs

Report this information for all failure messages. The order in which the messages are listed does not matter.

Sample Output

Below is a fragment of possible output from the program. Note that each section of output identifies one failure message and all of the associated data for that type of failure. Your program may follow this format.

Error from slot1@c149.chtc.wisc.edu: VMGAHP_ERR_INTERNAL

- Occurred 37 times from 04:40 to 05:23

- Affected 34 unique jobs

FAILED TO SEND INITIAL KEEP ALIVE TO OUR PARENT <128.105.244.224:9618?sock=3202_995c_3>

- Occurred 111 times from 04:29 to 04:29

- Affected 39 unique jobs

Input File

Download the sample data file logs.txt. Keep in mind that this is just one data file and that your program should work correctly on all files in the format described above.

Suggested Functions(optional):

You may use any combination of functions you wish to use to do this homework. The following is a suggested?outline for your program:

main(): Parse the input file name from the command line. Create a dictionary to save all the log entry. Pass the input file name and the dictionary to parse_file.Call the print_report.

parse_file():

Open the file and loop through all the error log and pass each one into?parse_record. And save the required log information into the dictionary.

parse_record():

Receive log information and exact job id and time.

print_report():

Loop through the dictionary and print all the information in required format.

Grading:

You will earn points as follows for each piece of functionality in your code:

Reading data source:

oProper error handling

oAllows user pass input file through command-line

oUse argparse as the parser for command-line options

Data parsing:

oUse efficiency way to read large file

oUse regular expression to exact job id and time from a string

oIgnores all no error log record

Data analysis:

oComputes the number of times that the error occurred correctly

oComputes the earliest and latest times that the error occurred correctly 5

oComputes the number of unique jobs that were affected correctly

Data output:

oData outputted in correct format

Total possible:

Code Quality:

Methods longer than 25 lines.

If you python file receive a negative score from pylint, you will receive 0 for that code. Other you lab score will be grading by the following formula, project score = (grading score) * ( (pylint score)/10 ).